So much misinformation going on here.
For open circuit diving where a diver incurs a decompression obligation and is under a ceiling, gas planning is done in such a way as to anticipate the failure/loss of one's breathing gas - this has little or nothing to do with how you measure it (AI/SPG/Rhythm Method).
AI as a "redundancy" is unnecessary because anything that could happen to render your primary means of measuring tank pressure useless would end the dive (and you know you have enough gas to exit the water, because that is a part of your plan).
If you want to dive AI then dive AI. In open water where there is no overhead, I don't see a real issue with it other than it having more o-rings and failure points that you don't really need. In an overhead there's a risk of impact and breakage causing an exposed HP port on the first stage. Again, not the end of the world, you'd just isolate and end the dive.
AI is one of those gimmicky things that divers who stay reasonably shallow and perform dives with low levels of complexity can safely rely on in lieu of really doing in depth dive planning - and that's OK for most vacation dives and most vacation divers.
Having one won't kill you, it also won't give you anything you don't already have. If you buy into the idea that we only take things into the water which we need to complete the dive then you'll likely not bother with it. If you like gadgets, you might. We don't use AI in our dive team because it's philosophically incompatible. if you can build a dive team with a different philosophy, more power to you.
I have six Shearwater computers (2 CCR controllers, 2 FC/PPO2 Monitors, 2 standalone bailout computers) and they have taken me to and from depths up to 137m safely. They're well built and reliable. I don't think they need AI, but if the vacation diver market will bare the price tag of the computer and Shearwater can incorporate it without compromising the functionality they've become so famous for offering then what's the harm?